use of group by, verifying my hunch

[sfullman]

I have a huge 1MM+ table. i want to pare it down. there are about 12 fields out of 50 that, if it weren't for them, would make an average of 5-10 records redundant (all values the same except for differences in these 12 fields).

Am I correct that the following are equivalent: (fields 1-4 are the repeated fields, and 5-17 are the unique ones)

1) select DISTINCT
field1,
field2,
field3,
field4
FROM
big_table

[i.e. the count of these]

2) select count(*)
FROM big_table
group by
field1,
field2,
field3,
field4

Thanks,
Sam Fullman

[Lars_Berg]

Yes.

[Sxooter]

To make the group by work the same you need to change it to:

select field1, field2, field3, field4
FROM big_table
group by
field1,
field2,
field3,
field4

but other that that they're pretty much the same.